International Journal on Electrical Engineering and Informatics - Volume 1, umber 1, 2009

Partial Discharge and Cross Interference Phenomena in a Three-phase Construction

Umar Khayam, Shinya Ohtsuka, Satoshi Matsumoto,
Masayuki Hikita

Kyushu Institute of Technology,

Department of Electrical and Electronic, and Computer Engineering,
1-1 Sensui-cho, Tobata-ku, Kitakyushu, 804-8550 Japan

Abstract: This paper deals with partial discharge measurement and partial discharge
characteristics on a three-phase construction. A simplified model of three-phase gas insulated
switchgear is used in the experiment to simulate three-phase electric field in three-phase
equipment. A particle was adhered on the different positions on the conductor and on the tank
to produce partial discharge. Partial discharge signals flowing on three phase conductors were
measured simultaneously with three high frequency current transformers clamped on the three-
phase lines through capacitive voltage divider. Partial discharge signals induced on the
enclosure were measured with the current transformer and the detecting impedance. The circuit
was calibrated and examined by injecting known charges from a pulse calibrator to some points
in the circuit. The calibration results indicate that measuring partial discharge signals only at
the enclosure or on one of the phases of three phase construction will result in the position
dependent sensitivity for partial discharges. Experimental results show that partial discharge
inception voltages of a particle adhered on different position differed depended on the particle
position. Phase of partial discharge inception voltage also depended on the particle position.
Experimental results show that partial discharge caused by a particle on one phase conductor
was induced and detected on the other phase conductors. Polarity and magnitude of discharge
current flowing on each phase depended on the particle position. Each particle at different
position generated discharge current which flowed on each phase so that the magnitude ratios
of the currents flowing on each phase were unique for each particle. These results were
discussed with considering electric field characteristic in three-phase construction and coupling
capacitance between three-phase conductors and between the conductors and the tank.
Index Terms : partial discharge, three-phase voltage, electric field, cross-interference

1. Introduction
DISCHARGE characteristics in single-phase equipments are a well-known and well-established art,
however, those in three-phase equipment such as three-phase gas insulated switchgear (GIS) and
three core belted power cable are far less understood. The investigation of partial discharge (PD)
characteristics under three phase electric field on three-phase equipment is needed because some
reasons. Firstly, there is a distinct difference in electric field characteristics between single-phase and
three-phase construction. The electric field in single-phase equipment is linear, while one in three-
phase equipment rotates and changes continuously [1-3]. It can be expected that the PD
characteristics in three-phase equipment are different from ones in single phase equipment. Further,
the particle movement inside the three-phase gas insulated system is influenced strongly by three-
phase electric field [2,4]. Secondly, for the reason of its compactness and low cost, the application of
three-phase equipment (three-phase in one tank), such as three-phase GIS and three-phase gas
insulated bus (GIB), has been increasing [5].

Corresponding Author: Umar Khayam, School of Electrical Engineering and Informatics, Institut Teknologi
Bandung, Jl. Ganesha 10 Bandung I DO ESIA, 40132 Telp./ Fax.: 62-22-2500995.
Email: umar@hv.ee.itb.ac.id
Thirdly, application of single-phase testing technique to three-phase equipment gives some erroneous
results [6,7]. Fourthly, three-phase equipment is energized by three-phase voltage on on-line PD
monitoring of medium and high voltage equipment [3]. It differs from common practice for off-line
test which is to energize only one phase.
Recently, the PD investigations on three phase equipment have been reported. The topics
included rotating electric field in three-phase equipment [1-3] and its effects on PD characteristics
[1,2] and particle movement on three-phase equipment [2,4], development of three-phase PD
detection using single-phase test voltage [6,7], and development of PD measurement technique on
three-phase equipment [8-16]. Fundamentals aspects of excitation and propagation of PD signal
in three phase power cable have been reported, however the PD excitation were only explained
with simulation [3]. PD characteristics, mainly explained from measurement on three-phase
construction, are still not adequate.
From these viewpoints, PD characteristics from a particle in a three-phase construction under
three-phase voltage are investigated. The measurement system is developed to observe discharge
characteristics under three-phase voltage. PD from a artificial particle adhered on the conductors and
on the tank were observed. In this paper, the PD occurrences and cross interference phenomena in
three phase construction are reported.

2. Experimental setup
The measurement system was developed to investigate discharge characteristics under three-
phase voltage. A simplified model of a three-phase GIS was used. It was composed of a tank model
150 mm in diameter, 300 mm in length, 2 mm in thickness, and three-phase conductors 25 mm in
diameter. The electrode system was arranged to simulate three-phase electric field in three-phase
equipment. The conductors were arranged in an isosceles triangle construction. The maximum
voltage of three-phase transformer used for the experiment was 52 kV. The dimensions of the tank
and the conductors were about one fifth of an actual 275 kV three-phase GIS. The layout and
dimensions of the model are shown in Figure 1. The experimental setup is shown in Figure 2.
The experiments were conducted in laboratory air. Three-phase AC voltage was supplied from a
three-phase 200 V/52 kV, 5 kVA corona free transformer through 20 k limited resistor. PD signal
flowing in each phase was measured by three current transformers with frequency band 10 kHz – 250
MHz, namely CTR, CTS, and CTT. They were clamped on the line connecting the coupling capacitors
and the capacitive dividers on R, S, and T phases, respectively. PD signal flowing from the tank to
ground was detected by CT, namely CTTANK, and 1 k detecting impedance (DI). The detecting
impedance is able to measure signal with frequency up to 300 MHz.
Applied voltage was measured with capacitive dividers consisting of 500 pF and 450 nF
capacitances. PD signals and applied voltage were observed simultaneously with a four-channel
oscilloscope.
340

75

R T
53

150 25 t2
S 150
300
R=20
 Figure 1. Layout of simplified model of three-phase GIS

Ck 3 GIS
PD
MODEL
CTR OSC

Cd VR
3 200V 200V / 52 kV 20k
r R R T
s S
S
t T
Ck PD PD
PT 13kV/100V CTS CT TANK
OSC OSC
Cd VS DI 1k

Ck = 500 pF PD
CTT VT OSC

Cd =450 nF

Figure 2. Equivalent circuit of experimental setup

3. Calibration and analysis of three-phase pd measurement system

The measurement system was calibrated to verify the relation between the reading of the PD
measurement system (mV) and injected charge (pC) and to verify sensitivity of the measurement
system. The calibration of PD measurement system was done by injecting positive and negative
known charges from a pulse calibrator to some injecting points as follows: phase to phase (R-S, R-T,
S-T) and phase to tank (R-Tank, S-Tank, T-Tank) respectively. The high frequency current was
measured with the PD measurement system and observed with an oscilloscope.

The examples of calibration results are shown in Figures 3 and 4. Figure 3 shows the response of
CT clamped on S phase (CTS) when charges are injected to some different points. It appears that CTS
detected charges injected to S phase. In addition, it also detected charges injected to another points,
such as R-T, R-tank, and T-tank. It means that charges injected to one phase were also induced to the
other phases and to the enclosure. It appears that the response of charges injected to S phase is higher
than that injected to the other points. It means that CTS has high sensitivity for discharge occurring
around S phase.

Figure 4 shows the response of CT clamped on the line connecting the tank and ground (CTTANK)
when charges are injected to some different points. It appears that the response of charges injected to
the tank (phase-tank or tank-phase) is higher than those injected to the other points (phase-phase). It
means that CTTANK has high sensitivity for discharge occurring between the conductor and the tank.

The results indicate some guidance for PD measurement on three-phase construction as follows.
a. PD occurring on one of three phase conductor induces charges to the other phases and the
enclosure.
b. Measuring PD current flowing in one of three-phase conductors is sensitive for discharges
which occur around the conductor (between the conductor and the tank, and between the
conductor and the other conductors), but less sensitive for discharges which occur
between the other two conductors and between the other conductors and the tank (Figure
3).
c. Measuring PD current flowing from the enclosure (tank) to ground is sensitive for
discharges which occur between each conductor and the tank, but less sensitive for
discharges which occur between the conductors (Figure 4).
d. Measuring PD current only at the enclosure of three phase construction or only at one of
the phases will result in position dependent sensitivity for PDs.
 The calibration results indicate that PD measurement must be conducted on three phases
simultaneously to obtain measurement results of PD occurring in entire region of three-phase
construction with high sensitivity,.

180
R-S
160 S-T
140 T-R
R-Tank
Voltage (mV)
120 S-Tank
100 T-Tank

80
60
40
20
0
0 100 200 300 400
Charge (pC)
Figure 3. Calibration results of CTS

180
R-S
160 S-T
140 T-R
Voltage (mV)

R-Tank
120 S-Tank
100 T-Tank

80
60
40
20
0
0 100 200 300 400
Charge (pC)
Figure 4. Calibration results of CTTANK

4. Experimental results
4.1 The Dependence of PDIV on particle position
Partial discharge from a particle placed at different positions on HV conductor and the tank
was measured. The particle positions and their notations are shown in Figure 5. The diameter of
a particle is 0.1 mm. The radius of particle tip is 0.05 mm. The particle length is 5 mm. PDIV
of these particles are shown in Figure 6. The results show that PDIV- is lower than PDIV+ for
the particles on HV conductor and on the contrary for the particles on the tank. PDIVs of a
particle adhered on different positions differed, depended on the particle position.
 E90

R90 T90
R T
S90 S S
S135 45

S S0
S270

Figure 5. Artificial particle positions and notations

PDIV vary between 4 kV and 6 kV for a particle adhered on the different positions on the
conductor and between 8 kV and 11 kV for a particle on the different positions on the tank.
PDIV of a particle on the tank is higher than PDIV of a particle on the conductor. PDIV of a
particle on the region between phases (R315, S90, and S135) is around 4 kV, lower than PDIV
of a particle on the other places.
12
PDIV N
10
PDIV P
8
PDIV (kV)

0
R0 R180 R270 R315 S90 S135 S180 S270 E0 E180 E270

Particle Position

Figure 6. PDIV of particles on the conductor and on the tank

4.2 The Phase dependence of pd occurrence on particle position

Figure 7 until 9 show experimental results of PD occurrence at PD inception voltage
(PDIV) from a particle on some different positions. Figure 7 and 8 show the examples of PD
occurrences during two cycles of applied voltage at PDIV for a particle adhered on the
conductor and a particle adhered on the tank respectively. It appears on Figure 7 that PDIV of
S45 particle occurred when phase of S voltage s is around 247O. On the other hand, PDIV of
S270 occurred when s = 270O. Then, it appears on Figure 8 that PDIV of particle on the tank
E180 and E270 occurred at s = 330O and s = 90O, respectively. The other experimental results
are summarized on Figure 9. It appears that phases of PDIV depended on the particle position.
 1.5

R S T 4
1.0

Applied Voltage (kV)

0.5 2

PD Current (mA)
0.0 0

-0.5

-1.0
PD -2

-4
-1.5
-15 -10 -5 0 5 10 15

Time (ms)
Particle S45, PDIV-, 4.7kV

1.5
4
1.0 R S T

Applied Voltage (kV)

2
PD Current (mA)

0.5

0.0 0

-0.5
-2

-1.0 PD -4
-1.5
-15 -10 -5 0 5 10 15

Time(ms)
Particle S270, PDIV-, 4.71 kV

Figure 7. PD occurrences on 2 cycles of applied voltage at PDIV (particle on the conductor)

15 8

S 6
10
PD R T
PD 4
Applied Voltage (kV)

5
PD Current (mA)

0 0

-2
-5
-4
-10
-6

-15 -8
-20 -10 0 10 20

Time(ms)

Particle E270, PD+, 7.76kV

15 8

R S T 6
10
Applied Voltage (kV)

4
5
PD Current (mA)

0 0

-5 PD -2

-4
-10
-6
-15
-15 -10 -5 0 5 10 15

Time (ms)
Particle E180, PDIV, 7.8kV

Figure 8. PD occurrences on 2 cycles of applied voltage at PDIV (particle on the tank)

 S T R
E270

Applied Voltage
E0 E180

PD Pulse
T0 T225 R180 S45 S90 S135

R315 S270

0 30 60 90 120 150 180 210 240 270 300 330 360

Phase Angle of S Phase Voltage

Figure 9. Resume of phase of applied voltage at PDIV for a particle on some different
positions

4.3 Cross interference phenomena

Figure 10 shows waveform of discharge current flowing on each phase for S90 particle. PD
currents measured by CTR, CTS, CTT, and CTTANK were 0.40mA, 0,88mA, 0.48mA, and
0.25mA, respectively. It appears that discharge caused by particle on S phase caused currents to
flow on R, S, and T phases and to the enclosure. It means that the discharge occurring on one
phase was induced and detected on the other phases. The charge was also induced to the tank.

For further analysis the magnitude of PD currents flowing on each phase measured by CT is
classified. The magnitude of PD current flowing on each phase and the enclosure are
normalized. Then, the maximum of normalized currents flowing on each phase, further is called
as magnitude ratio (A), were determined. The magnitude ratio is classified as follows: low (L)
for A <0.4; medium (M) for 0.4 A<0.7; and high (H) for A 0.7. Based on the classification,
PD pulses from S90 particle as shown in Figure 10 result in low and negative current detected
by CTTANK, medium and negative current detected by CTR and CTT, high and positive current
measured by CTS. Example of the other results is summarized in Table 1.
-3
x10

5 DI
0
-5
- 0.25mA 0.28I L CTTANK
0.5
0.0
-0.5
PD Current (mA)

- 0.40mA 0.45I M CTR

0.5
0.0
-0.5
+ 0.88mA I H CTS
0.5
0.0
-0.5
- 0.48mA 0.55I M CTT
0.5
0.0
-0.5

-0.1 0.0 0.1 0.2 0.3 0.4 0.5

Time (us)

Figure 10. PD Current waveform measured for S90 particle

 Table 1 Polarity and magnitude level of discharge currents measured for particles on S
conductor

Particle PD Discharge Current

Position Polarity CTTANK CTR CTS CTT
S90 PD- L- M- H+ M-
S90 PD+ L+ M+ H- M+
S45 PD- L- L- H+ H-
S45 PD+ L+ L+ H- H+
S270 PD- H- L- H+ L-
S270 PD+ H+ L+ H- L+
S135 PD- L- H- H+ L-
S135 PD+ L+ H+ H- L+

Even though it is not shown here, the repetition of the experiment on the same condition
showed that the magnitude ratio is constant and independent of magnitude of PD current pulse.
These results indicate that the ratio of magnitude of PD current flowing on each phase provides
information of the particle position on the cross section of three phase construction.

5. Discussion
5.1 Electric field inside the three-phase construction and pd occurrences
The PDIV characteristics are discussed first. The results show that PDIV- is lower than
PDIV+ for the particles on HV conductor and on the contrary for the particles on the tank.
Although these results are natural phenomena, let us discuss these results since it is important to
explain the dependence of phase of applied voltage at PD occurrence on the particle position.
Positive PD is generated by an electron initiated by collision detachment of negative ions in
high electric field region. Thus, positive PD generation depends on whether or not negative
ions exist around the particle tip. The negative ions at positive PD inception phase can be
derived from PD in the previous cycle and or cosmic rays. The high field region initiating PD
is so restricted that, even under continuous three-phase AC voltage application, the generation
of an initial electron would be necessary condition for PD initiation. On the other hand,
negative PD is generated by an initial electron derived from the field emission from the
electrode surface. Thus, the generation of an initial electron for negative PD depends only on
the electric field strength on the electrode surface. Therefore, an initial electron can be
generated more easily for negative PD than for positive PD.

In the case of the particle on S conductor, the polarity of particle tip is negative when the
polarity of S voltage is negative. Therefore, the first PD occurs in the negative half cycle of S
voltage. In the case of the particle is on the tank and face to S conductor, the polarity of the
particle tip is negative when the polarity of S voltage is positive. Therefore, the first PD occurs
in the positive half cycle of S voltage. Because the particle adhered on the tank needs higher
voltage to attain the same level of electric field than particle adhered on the conductor, PDIV of
particle on the tank is higher than PDIV of particle on HV conductor.

PDIV is determined by PDIV criterion which depends on effective ionization coefficient,

[17,18]. PDIV criterion is given by Schumann equation as follows.

(1)

where = I- . I is Townsend’s first ionization coefficient and is the attachment coefficient,

and both are functions of the electric field, atmospheric pressure, and temperature. X is the
distance from the highly electrode surface and the point near electrode surface where I= . It is
sometimes called the critical avalanche length or the boundary of the corona layer. K = 20 for
air at 1 atm [18].

PDIV therefore depends on the electric field on the tip and the field distribution along the
axis of the particle since is a function of the electric field intensity. For air [18]:

/ p = 22 (E/p – 0.244)2 for 0.244 < E/p < 0.5 (2)

where E is electric field intensity in kV/cm, while p is air pressure in kPa. Because the electric
field on the particle tip is much higher than on the other part, PDIV is influenced mainly by the
electric field on the particle tip.

Let’s observe the electric field characteristics inside the three-phase construction. Figure 11
shows periodic change of absolute electric field on the particle tips during one cycle of S
applied voltage at different positions (R315, S45, E270, and T0). The electric field on each
particle tip was calculated at 1 kV applied voltage for the comparison convenience. The electric
field calculation point is determined so that the electric field at the point at PDIV is equal to
critical electric field Ecr in air at atmospheric pressure ( Ecr = 3 kV/mm ). The critical electric
field is the field at which the ionization begin to take place if I= .

A B C D
0.8

0.7
Electric Field (kV/mm)

0.6

0.5
R315
0.4 S45
T0
0.3
E270
0.2

0.1

0.0
0 30 60 90 120 150 180 210 240 270 300 330 360

Phase of S Voltage (0)

Figure 11. Periodic change of electric field on the particle tips

As shown in Figure 11 at certain position (x,y) inside the three phase construction the
electric field is rotating with varying magnitude, i.e sinusoidal in two perpendicular directions
without necessarily coinciding zero crossings. The electric field stress on each point inside the
construction varies with the phase of the applied voltage. Every a half cycle the field achieves
the minimum and the maximum values. Every a half cycle the pattern of the electric field stress
is repetitive. It appears in Figure 11 that the value of the maximum electric field depends on the
particle position. It is reasonable because the gap distance and the electrode configuration of
the three-phase construction depend on the particle position.
Figure 12 shows the maximum electric field on the particle tip of a particle on various positions
at 1 kV applied voltage. It appears that the electric field of the particle on the conductor face to
the other conductor (phase-phase region) is higher than that on the other place. Because PDIV
depends on the electric field, these electric field calculations explain reasonably the
experimental results: PDIV of particles with the same size depend on the particle position; and
PDIV of the particles on the conductor facing to the other conductor is higher than that on the
other place.

It is also shown in Figure 12 that the maximum electric field on the particle tip on different
positions occurs at different phase of applied voltage. For example, the maximum electric field
on the tip of S45 particle occurs at the phase of S voltage S is 67o and 247o, while one of E270
particle occurs at S = 90o and S = 270o. Because PD is initiated at the maximum electric field,
these electric field characteristics result in the phase dependence of PD occurrence at PDIV on
the particle position.

Figure 13 summarizes phase angle of S applied voltage at maximum electric field for a
particle on the different position on R, S, and T conductors and on the tank. In general, PD of a
particle on one phase conductor occurs on the peak of applied voltage at the related phase. The
effects of the other phases take a place when the particle faces to another conductor, resulting in
the shifting of the phase of maximum electric field and PDIV from the peak of applied voltage.
These calculation results explain the experimental results summarized on Figure 9. For
example, negative PDIV of S45 particle occurred at S = 247o, while negative PDIV of E270
particle occurred at S =90o. Note that negative PDIV is easier to occur than positive PDIV for
a particle adhered on the conductor and on the contrary for a particle adhered on the tank.

0.8
Electric Field on the Particle Tip

0.7

0.6

0.5
(kV/mm)

0.4

0.3

0.2

0.1

0
R0 R180 R270 R315 S90 S135 S180 S270 E0 E180 E270

Particle Position

Figure 12. Electric field on the particle tip at 1 kV applied voltage

 Phase angle of S phase voltage
360
330
300
270
240

at PDIV
210
180
150
120
90
60
30
0

R0
R90
R180

R315
S0

S90
S135

T90
T180
T225
T270
E0
E90
E180
R270

S45

S180
S270
T0

E270
Particle Position

Figure 13. Phase of maximum electric field for a particle on the different position

5.2 Cross interference phenomena

The experimental results revealed that discharge from particle on one phase conductor
caused currents to flow on the other phases, not only on the phase where the particle exists. The
equivalent circuit of three-phase configuration with S90 particle and the measurement system
are shown in Figure 14 to explain the phenomena. Partial discharge resulting from S90 particle
is described as a pulse source. Insulation regions between the tank and each conductor and
between conductors are described as their equivalent coupling capacitors; those are CR, CS, CT
and CRS CST and CRT. By LCR meter, the capacitances are estimated to be CR=9.5 pF, CS=7.9 pF,
CT=9.6pF, CRS=5.3 pF, CRT=3.7 pF, CST=6.1 pF.

Let us analyze the circuit. If negative discharge occurs in S90 particle, a small amount of
charge is induced to R and T phase. A large amount of charge flow in S phase conductor to
ground (path S). Since the current flows toward positive sign of CTS, it is measured as positive
current by CTS. Then the current is divided into three directions: R conductor (path R), T
conductor (path T), and the tank (path TANK). Since they flow toward negative sign of each
CT, it is measured as negative current by CTR, CTT, and CTTANK. The currents are transmitted
from R and T conductor to tank, and then from tank to S phase conductor, because of the
presence of coupling capacitances between the two conductors and between conductor and the
tank. These discharge current path are depicted in Figure 14.

The discharge current flowing in each phase was estimated based on the equivalent circuit
and dominant frequency of discharge current. The frequency was estimated based on the rise
time of PD current waveforms measured for particles attached on different positions. Average
rise time (tr) of PD current was 17.5 ns so that the dominant frequency of PD current can be
determined by the relation f = 0.35/tr, giving f = 20 MHz.

If the current flowing in path S is I (high), the currents flowing in R, T, and tank paths are
0.46I (medium), 0.53I (medium), and 0.01I (low), respectively. These calculation results agree
with the magnitude of discharge current resulting from the measurement in the term of
magnitude level, although there are small differences in values.
 Figure 14. Equivalent circuit and path of PD current of S90 particle

6. Conclusions
PD characteristics in three-phase construction under three-phase voltage were investigated.
The measurement system for observing discharge characteristics in three-phase construction
was developed. PD from a particle on the different positions on the conductor and on the tank
was observed. The conclusions are as follows.

1. Measuring PD signals only at the enclosure or on one of the phases of three phase
construction will result in the position dependent sensitivity for partial discharges.
2. Partial discharge inception voltage of a particle adhered on different position differs
depending on the particle position.
3. Phase where PD occurs at PDIV depends on particle position.
4. PD caused by a particle on one phase conductor is induced and detected on the other
phase conductors.
5. Magnitude ratios of PD current flowing on each phase are unique for each particle on
different positions. The ratio is constant and independent of magnitude of PD current
pulse.
6. Polarity of PD current flowing on each phase depends on particle position.

Calibration and measurement results indicate that PD measurement on three phase

construction must be conducted on three phases simultaneously to obtain measurement result of
PD occurring in entire region with high sensitivity and to obtain information of the location of
PD source. Experimental results indicate that phase of applied voltage at PDIV and magnitude
ratio of discharge current flowing on each phase provides information of the particle position
on the cross section of three phase construction. Therefore, they can be considered as
parameters for estimation of particle position on the cross section of three phase equipment.

References
[1] Umar Khayam, T. Ishitobi, S. Ohtsuka, S. Matsumoto, M. Hikita: “Effect of Elliptical
Nature of Rotating Electric Field on Partial Discharge Pattern in a Three-phase
Construction”, IEEJ Transaction on Fundamentals and Materials, Vol. 127, No.9, 2007,
pp. 524-530.
[2] S. Yanabu, H. Okubo, S. Matsumoto,“Metallic Particle Motion in 3-Phase SF6 Gas
Insulated Bus”, IEEE Transactions on Power Delivery Vol. PWRD-2, No. 1, January
1987.
[3] Wielen, E.F. Steennis, and P.A.A.F Wouters,”Fundamental aspects of excitation and
propagation of on-line PD signals in three phase medium voltage cable systems,”, IEEE
Trans. on DEI, vol 10, no4, pp. 678-688, Aug. 2003.
[4] S. Yanabu, Y Murayama, and S. Matsumoto,”SF6 Insulation and its Application to HV
Equipment”, IEEE Transactions on Electrical Insulation, Vol. 26, No.3, pp 358-366,
1991.
[5] G.V.N. Kumar, J.Amarnath, B.P. Singh, K.D. Srivastava,”Electric Field Effect on Metallic
Particle Contamination in a Common Enclosure Gas Insulated Busduct”, IEEE Trans. on
DEI, Vol 14, No. 2, pp. 334-340, Apr. 2007.
[6] Kreuger, F.H, Samir Shihab, “Partial Discharge Measurement on Three-core Belted Power
Cables”, IEEE Transactions on Power Delivery, Vol. 4, No. 2, April 1989
[7] E. Harkink, F.H. Kreuger, P.H.F. Morshuis: “Partial Discharges in Three-core Belted
Power Cables”, IEEE Transactions on Electrical Insulation, Vol. 24, No. 4, pp. 591-598,
Aug. 1989.
[8] R. Heinrich, S. Schaper, W. Kalkner, R. Plath, A. Bethge,”Synchronous Three-phase PD
Detection on Rotating Machones”, XIIIth ISH, etherlands 2003.
[9] S.Schaper, W. Kalkner, R.Plath, H. Emanuel,”Synchronous Multi Terminal PD Detection
on Power Transformer”, XIIIth ISH, etherlands 2003.
[10] W. Kalkner, A. Obralic, M. Kaufhold, R. Plath,”Investigation of PD-Pulse Propagation
and Coupling between Three-phases of a Stator of a Rotating Machine”, XIVth ISH, Paper
B-44, China, 2005.
[11] W. Kalkner, A. Obralic, A. Bethge, R. Plath,”Synchronous 3-Phase Partial Discharge
Detection on Rotating Machines”, XIVth ISH, Paper G-042, China, 2005.
[12] A.M. Jafari, A. Akbari,”Multi-Source Partial Discharge Diagnosis in Three-phase High
Voltage Devices”, XIVth ISH, Paper G-051, China, 2005.
[13] Ronald Plath,”Multi-channel PD Measurement”, XIVth ISH, Paper J-04, China, 2005.
[14] Akbari, A.M. Jafari, M. Kharezi, “Partial Discharge Localisation in Power Transformers
with Delta Connected Windings”, XIVth ISH, Paper G-052, China, 2005.
[15] A.M. Jafari, A. Akbari,”Multi-Source Partial Discharge Diagnosis in Three-phase High
Voltage Devices”, XIVth ISH, Paper G-051, China, 2005.
[16] Ronald Plath,”Multi-channel PD Measurement”, XIVth ISH, Paper J-04, China, 2005.
[17] D.B. Philips, R.G. Olsen, P.D. Pedrow,”Corona Onset as a Design Optimization Criterion
for High Voltage Hardware”, IEEE Transaction on Dielectrics and Electrical Insulation
Vol. 7 No. 6 pp. 744-751, December 2000.
[18] N.H. Malik and A.H. Qureshi,”Calculation of Discharge Inception Voltage in SF6-N2
Mixtures”, IEEE Transaction on Electrical Insulation Vol. EI-14 No.2 pp. 70-76, 1979.

Umar Khayam was born in Indonesia in 1975. He received B.Eng. with honor (cum laude) and
M.Eng. degrees in electrical engineering from Institut Teknologi Bandung
(ITB), Indonesia, in 1998 and 2000, respectively. He worked as a researcher
and lecturer assistant at High Voltage and High Current Engineering
Laboratory, ITB, from 1999 to 2004. He received Doctor Degree in Electrical Engineering
from Kyushu Institute of Technology (KIT), Japan in 2008. He was a researcher at Hikita
Laboratory, KIT, Japan, during 2008-2009. Since 2008, he is a Lecturer at School of Electrical
Engineering and Informatics ITB, Indonesia. His research interests include the partial discharge
measurement and phenomena in electric power apparatus. He is a member of IEEJ since 2005.
Dr. Umar Khayam can be contacted at School of Electrical Engineering and Informatics ITB, Jl.
Ganesha 10 Bandung, Indonesia 40132 or at umar@hv.ee.itb.ac.id.

Shinya Ohtsuka was born in 1971. He received his B. S. and M. S. degrees

from Kyushu University, Japan, in 1994, 1996, respectively. He finished
his Doctor Course of the Deprt. of Electrical and Electronic Systems
Engineering at Kyushu University in 1998. He was a research fellow
(DC1) of the Japan Society for the Promotion of Science (JSPS) from 1996
to 1998, then became a JSPS Research Fellow (PD) in 1999. Since 1999,
he has been an Assistant Professor at Kyushu Institute of Technology,
Japan and promoted to Associate Professor at the same university in 2006.
His research interests include the insulation properties of environmentally-
benign gas as SF6 substitute, PD detection technique for insulation diagnosis of power
apparatus and the superconductivity engineering. He has a Doctor Degree of Engineering. Dr.
Ohtsuka is a member of the IEEJ, the IEEDJ and the Cryogenic Society of Japan.

Satoshi Matsumoto (Member) was born in 1955. He received the Ph.D in 1984 in electrical
engineering from the University of Tokyo, and joined Toshiba Corporation.
He was visiting Professor at Kyushu Institute of Technology from 2003-
2006. He is presently visiting Professor at Shibaura Institute of
Technology. He is a member of the IEEJ, and Senior Member of IEEE.

Masayuki Hikita (M’96-SM’00) was born in 1953. He received B. S., M. S.

and Dr. degrees in electrical engineering from Nagoya University of Japan,
in 1977, 1979, and 1982, respectively. He was an Assistant, a Lecturer, an
Associate Professor at Nagoya University in 1982, 1989, and 1992,
respectively. Since 1996, he has been a professor of the Department of
Electrical Engineering, Kyushu Institute of Technology. He was a Visit
Scientist at the High Voltage Laboratory in MIT, USA, from August 1985 to
July 1987. Dr. Hikita has recently been interested in research on
development of diagnostic technique of electric power equipment. He is a member of the Japan
Society of Applied Physics and the IEEJ. He is now a senior member of the IEEE.